What's accessing my hard disk??? (Help Please!!)

[stevegong]

Normally when you put your computer to just sit there and do nothing, nothing should be accessing the hard disk. My hard disk clicks about once or twice a minute and this was preventing my hard disk to spin down whenever possible.

Further inspection in Activity Monitor shows me that in fact, the disk activity does spike once in a while, and it's red meaning it's not a user process.

I just reformatted my hdd today and reinstalled a fresh copy of panther and it still didn't solve the problem. I ran the hardware test and it says my hardware has no problems. I reset the power manager, booted into open firmware and typed reset-all, and reset the pram.

still the proble is there.


What's going on?

Please help, it's driving me nuts!!!!!!!!!!

[curiousuburb]

if you have less than ample RAM, it might be virtual memory pageouts

not enough information to diagnose further

[stevegong]

I maxed out on ram on my machine, 640 megs total.

I'd highly doubt it's the ram.

Do you have any other ideas?

Thanks btw, I appreciate any comments or ideas.

[DMBand0026]

You got the firewall on on that thing? It's possible (however unlikely) that someone is ssh'ing into your machine or some other remote login process.

[Mac+]

First - you might get more traffic if this was in Genius Bar.

However, can you tell us a bit more please...
How many apps are running at once?
When was the last time you either restarted the machine, repaired the permissions or in general gave the machine a "clean start"?
How much space do you actually have available on your HD?

I would have guessed that it had something to do with paging out. My 20GB HD used to click away incessantly, especially when I only had about 1GB or less of space available on it! Finally, though, I bought an external FW drive and archived a lot of my files and data. This then cleared up my HD space and I'm happy to say that the clicking noise from my HD is almost a non-event these days! :)

[SonOfSylvanus]

Normally when you put your computer to just sit there and do nothing, nothing should be accessing the hard disk. My hard disk clicks about once or twice a minute and this was preventing my hard disk to spin down whenever possible.

Further inspection in Activity Monitor shows me that in fact, the disk activity does spike once in a while, and it's red meaning it's not a user process.

I reformatted my hdd and reinstalled a fresh copy of panther and it still didn't solve the problem. I ran the hardware test and it says my hardware has no problems.


What's going on?

Maybe it isn't the HD clicking at all. My PB clicks aswell (even when idle), but I thought it might be the graphics card (for some reason).

[stevegong]

First - you might get more traffic if this was in Genius Bar.

However, can you tell us a bit more please...
How many apps are running at once?
When was the last time you either restarted the machine, repaired the permissions or in general gave the machine a "clean start"?
How much space do you actually have available on your HD?
! :)


I'm running only iChat but even if I quit every single programme, something still accesses the hard disk. (as stated before it's a non user process that's accessing the hdd)

Well, I reformatted the entire hard disk and make a clean reinstall of the os yesterday, so... (nothnig to do with restarting or repairing permissions)

I have 53 gigs available space.


And it's definitely the hard disk clicking because it syncronises with a peak in red in the Activity monitor disk access.

No firewall and definitely no one's trying to access me. (and the problem persists even if I unplug everything)

[Defiant]

What is the name of the process that spikes up when the access takes place?

[stevegong]

What is the name of the process that spikes up when the access takes place?

I don't know how to tell. A couple of processes are always spiking up and down so I don't know which one it is. All I know is that on the graphic meter, I see a red spike up every once in a while.

How can I tell which process it is? Can you please tell me how to do that?

Thanks a lot.

[stevegong]

Another thing I noticed recently is that I lost the ability to double tap to open a file or folder using the trackpad. I can single tap, drag lock, but not double tap to open something. If I double tap on the bar of a window though, it does recognise it and minimises.

The trackpad goes haywire say, maybe once a month? Totally crazy, and then after resetting the power manager it goes back to normal.


I'm thinking it might be a broken power manager? Or maybe the 2 problems are totally unrelated?

My machine I got in october, so it's not even that old, and all these problems are quite worrying.

[Brad]

Moving to the Genius Bar. :)

Here's a terminal command you can use to monitor disk access:

fs_usage

You'll have to use sudo as an admin user to use it (ie. use the combined command sudo fs_usage). Be sure to close and quit *everything* that you know isn't accessing the disk beforehand because a lot of entries may fly by, making the output unreadable.

[stevegong]

Here's a terminal command you can use to monitor disk access:

fs_usage

You'll have to use sudo as an admin user to use it (ie. use the combined command sudo fs_usage). Be sure to close and quit *everything* that you know isn't accessing the disk beforehand because a lot of entries may fly by, making the output unreadable.


hmm.... this is what i get:

Last login: Sun May 30 03:58:22 on ttyp1
Welcome to Darwin!
Stephen-Gongs-Computer:~ Stephen$ sudo fs_usage
Password:
fs_usage: the trace facility is currently in use...
fs_usage, sc_usage, and latency use this feature.

what does it mean?

[Brad]

Currently in use? Hmm... this could get interesting! :)

Have you quit everything? Including Activity Monitor?
Do you have any haxies or Konfab widgets or other third-party addons running?

[Defiant]

Just for fun, I tried this too before. Worked fine, no problem. I saw all disk access flying by, pretty spiffy. Then I closed the window and did some other stuff. Later, I decided I wanted to see it again. No dice. Got the already in use error message.
So I had a look at top, and there it was, still in use. After I killed it, it worked fine again. But as soon as I closed it, and tried again, it refused to work.

That means that this fs_usage process runs rampant, even after closing the terminal and all. So, if you run into this, do the following:
Open terminal, type 'top'
look out for a process called 'fs_usage', and note yourself the PID of it
press q (to quit top)
enter 'sudo kill PID' where PID is the number you noted before
now everything is fine again, run fs_usage by typing 'sudo fs_usage'
HTH with your problem.

[HOM]

Could OSX be writing to the Journal?

[Mac+]

Steve - running only one app and having recently reformatted the HD, in addition to having access to a fair amount of space does sort of knock out the "page out" suggestion. Well, I'm fresh out of ideas! :lol:

However, looking at the posts above, I'd say you're in expert hands now ;) ... when Brad or Defiant or Kickaha or LoCash comes along, I usually just sit back and watch and try to pick up/understand as much as I can.

AI.org Tech Support - picking up steam ... I love it! :)

PS - once my power cable was faulty and this caused intermittent clicking on the HD (or it could have been the internal power supply)... is there anything there to check out?

[stevegong]

oh, actually, the reason I had fs_usage already running was because I was checking in an old thread and they suggested to check that. but that thread didn't go anywhere and I think the reason their hard disks didn't spin down was because of some MS office scheduler.

So what am I supposed to look for here in the fs_usage? it keeps on scrolling and occasionally something pops up.


I'm starting to think this is a hardware problem, even though it's hard to imagine it that way.

BTW, I installed panther on my external fw hdd too, and same thing there, just that when I boot from the external, it is the external (startup volume) that has trouble spinnig down, whereas my internal (slave) spins down ok.

[_Ω_]

Do you sync your machine with anything?

[stevegong]

Do you sync your machine with anything?


Well I don't think that's really relevant because I'm talking about a complete reformat and clean install with no third party apps installed (not even iphoto) and everythnig unplugged too.

[_Ω_]

Reason why I asked is I have heard sometimes the synch process can get stuck when you synch to your iDisk. So even though you are not connected your machine is "looking" for the connection.

But I will keep out of this thread from now on.....

;)

[curiousuburb]

ok... let's try asking the obvious...

do you have Filevault enabled ?
(this would account for regular activity as it encrypts/decrypts on the fly)

is network activity simultaneous with disk activity ?
network threads (including rendezvous announce/discovery) process higher than many others and can account for some drive activity at times when no indicated application seems to be running.

P2P running?
Acquisition will stream info about your shared files to other users (ultrapeer settings or not), and seems to occasionally check not only for incomplete files getting moved into the active directory, but also scans subdirectories for newly shared files... you'd probably see it as a process, but if all you notice is network access and port 63xx traffic, I'd suspect P2P.

Mail running?
Apple's Mail program seems to occasionally check the address book and index mailboxes (particularly when iChat is open), despite otherwise running as a background process.

Scheduled events?
Software Update calling home? any other apps checking their mothership for updates (Acrobat, etc). some have a specific menu or preference item for frequency of update check...

once we can categorically eliminate these as causes, the troubleshooting may get easier

[stevegong]

ok... let's try asking the obvious...

do you have Filevault enabled ?
(this would account for regular activity as it encrypts/decrypts on the fly)

is network activity simultaneous with disk activity ?
network threads (including rendezvous announce/discovery) process higher than many others and can account for some drive activity at times when no indicated application seems to be running.

P2P running?
Acquisition will stream info about your shared files to other users (ultrapeer settings or not), and seems to occasionally check not only for incomplete files getting moved into the active directory, but also scans subdirectories for newly shared files... you'd probably see it as a process, but if all you notice is network access and port 63xx traffic, I'd suspect P2P.

Mail running?
Apple's Mail program seems to occasionally check the address book and index mailboxes (particularly when iChat is open), despite otherwise running as a background process.

Scheduled events?
Software Update calling home? any other apps checking their mothership for updates (Acrobat, etc). some have a specific menu or preference item for frequency of update check...

once we can categorically eliminate these as causes, the troubleshooting may get easier


Well, we can definitely assume these causes have been eliminated since I'm running NO programmes and I've never used file vault.

So what I tried is actually log out and in the login window, I waited and the hard disk did indeed spin down after 10 minutes, only to spin up again shortly thereafter (1-2 mins).


The problem is it can't sustain the hdd spun down.


Thanks guys for all the replies. I'm hoping to solve this problem since I won't be able to go to an apple store to have it fixed until late august :confused: :mad:

[stevegong]

UPDATE

So I ran the fs_usage with everything not running and sat there for about 10 minutes. I think I found what's creating the problem, but can someone tell me what it is?

21:19:54 WrData 0.000286 W update
21:19:54 sync 0.303327 W update
21:19:54 WrMeta[async] 0.002716 W update
21:19:54 WrMeta[async] 0.002659 W update
21:19:54 WrMeta[async] 0.002779 W update


21:22:24 WrData 0.000279 W update
21:22:24 sync 0.306732 W update
21:22:24 WrMeta[async] 0.002729 W update
21:22:24 WrMeta[async] 0.002909 W update
21:22:24 WrMeta[async] 0.002849 W update


21:24:24 CACHE_HIT 0.000007 WindowServer
21:24:24 WrData 0.000288 W update
21:24:24 WrMeta[async] 0.000995 W update
21:24:24 WrMeta[async] 0.001102 W update
21:24:24 WrMeta[async] 0.001395 W update
21:24:24 WrMeta[async] 0.001443 W update
21:24:24 WrMeta[async] 0.001454 W update
21:24:24 WrMeta[async] 0.001650 W update
21:24:24 WrMeta[async] 0.001748 W update
21:24:24 WrMeta[async] 0.001903 W update
21:24:24 WrMeta[async] 0.002305 W update
21:24:24 sync 0.312174 W update
21:24:24 WrMeta[async] 0.004712 W update
21:24:24 WrMeta[async] 0.004925 W update
21:24:24 WrMeta[async] 0.003304 W update
21:24:24 WrMeta[async] 0.003361 W update
21:24:24 WrMeta[async] 0.003525 W update
21:24:24 WrMeta[async] 0.003757 W update
21:24:24 WrMeta[async] 0.003742 W update
21:24:24 WrMeta[async] 0.003890 W update


21:28:54 stat /usr/share/zoneinfo/Europe/Rome 0.000070 notifyd
21:28:54 lstat private/etc/localtime 0.000040 notifyd

21:28:54 WrData 0.000266 W update
21:28:54 WrMeta[async] 0.001207 W update
21:28:54 WrMeta[async] 0.000938 W update
21:28:54 WrMeta[async] 0.001026 W update
21:28:54 WrMeta[async] 0.001103 W update
21:28:54 WrMeta[async] 0.001278 W update
21:28:54 WrMeta[async] 0.001462 W update
21:28:54 WrMeta[async] 0.001663 W update
21:28:54 WrMeta[async] 0.001857 W update
21:28:54 WrMeta[async] 0.002143 W update
21:28:54 sync 0.280085 W update
21:28:54 WrMeta[async] 0.002977 W update
21:28:54 WrMeta[async] 0.002937 W update
21:28:54 WrMeta[async] 0.002950 W update
21:28:54 WrMeta[async] 0.003119 W update
21:28:54 WrMeta[async] 0.003431 W update
21:28:54 WrMeta[async] 0.003649 W update
21:28:54 WrMeta[async] 0.003908 W update


21:27:00 stat tabs 0.000064 cron
21:27:00 stat private/etc/crontab 0.000042 cron


21:22:40 select 1.087063 W cupsd

21:23:54 stat /usr/share/zoneinfo/Europe/Rome 0.000069 notifyd
21:23:54 lstat private/etc/localtime 0.000040 notifyd


It looks like a combination of cron, the sync network time, and the one that comes out most is this W update thing. The W update thing happens when I hear the hard disk tick so I think it's probably W update. but what is it?

The code here is in cronological order and I've taken out the cache hits.


Any ideas? I appreciate it, really.

[stevegong]

Will no one help a lost soul????? :(

[thuh Freak]

cron, and windowserver i know. cron is a scheduler, which runs programs...on a schedule. windowserver manages all the open cocoa programs & windows and everything. W is a bit more vexing. i looked up the W program, and more strangely, there is a miniscule w program (i couldn't find anything about 'W'; capitalization doesn't matter on most macosx things, but sometimes it does on the lowerlevel stuff). 'w' lists the current users, and idle times and stuff. and it reads it from a file /var/run/utmp. so, some program is probably calling w, to check on what users are connected, every so often.

i have an idea of how we can investigate this, but its a bit dirty. and i must warn you before we try. if you don't make a backup, and replace it after we're finished with the test, bad things might happen. be prepared for some down and dirtiness. what i propose you try to do is move 'w' out of the way, and hopefully whatever program that is trying to use it will give somekind of useable error message. then you can see who is using it and we can decide what to do with that program. note: in trying this, we may break unrelatd programs. don't be overly alarmed, we'll replace everything when we're done.

so here's what you can do: open up terminal (from /Applications/Utilities). and run this command:
sudo mv /usr/bin/w /usr/bin/w.OUTOFTHEWAY
(you should copy and paste the above, to make sure you get it exact)

now, if another program tries to call 'w', it will not be able to find it. with any luck, the calling program will spit an error message somewhere. try leaving the computer alone for a few minutes/hours, see if the hd spins down, and more importantly make note of any popup errors that come along. an error might also be put in the console (i forget where those logs are saved, but the console app i think is in /Applications/Utilties too). report back any new errors or relevant console logs.

to fix 'w', do the following in terminal:
sudo mv /usr/bin/w.OUTOFTHEWAY /usr/bin/w
(again, c+p the above to get it exact)

[stevegong]

ok thuh Freak, did exactly as you said. then I decided to reboot so I could make sure this w process is not running. It did not fix the problem as my hdd never spin down. Then I decided to do a fs_usage:

I get this just about every second:

03:33:23 select 1.000047 W cupsd

here's a snippet. As you can see, sometimes, the W update is followed by the cron scheduler. I don't know if they are in fact related though, cause most of the time W update shows up without cron.


03:34:00 stat tabs 0.000060 cron
03:34:00 stat private/etc/crontab 0.000036 cron
03:34:00 CACHE_HIT 0.000082 WindowServer
03:34:00 CACHE_HIT 0.000021 WindowServer
03:34:00 CACHE_HIT 0.000022 WindowServer
03:34:00 CACHE_HIT 0.000010 WindowServer
03:34:00 CACHE_HIT 0.000013 WindowServer
03:34:00 CACHE_HIT 0.000009 WindowServer
03:34:00 CACHE_HIT 0.000023 WindowServer
03:34:00 CACHE_HIT 0.000010 WindowServer
03:34:00 CACHE_HIT 0.000008 WindowServer
03:34:00 CACHE_HIT 0.000007 WindowServer
03:34:00 CACHE_HIT 0.000007 WindowServer
03:34:00 CACHE_HIT 0.000007 WindowServer
03:34:01 select 1.000036 W cupsd
03:34:00 CACHE_HIT 0.000082 WindowServer
03:34:00 CACHE_HIT 0.000023 WindowServer
03:34:00 CACHE_HIT 0.000022 WindowServer
03:34:00 CACHE_HIT 0.000010 WindowServer
03:34:00 CACHE_HIT 0.000013 WindowServer
03:34:00 CACHE_HIT 0.000010 WindowServer
03:34:00 CACHE_HIT 0.000020 WindowServer
03:34:00 CACHE_HIT 0.000008 WindowServer
03:34:00 CACHE_HIT 0.000007 WindowServer
03:34:00 CACHE_HIT 0.000007 WindowServer
03:34:00 CACHE_HIT 0.000007 WindowServer
03:34:00 CACHE_HIT 0.000007 WindowServer
03:34:00 CACHE_HIT 0.000120 WindowServer
03:34:00 CACHE_HIT 0.000023 WindowServer
03:34:00 WrData 0.003416 W update
03:34:01 CACHE_HIT 0.000060 WindowServer
03:34:01 CACHE_HIT 0.000019 WindowServer
03:34:01 CACHE_HIT 0.000020 WindowServer
03:34:01 CACHE_HIT 0.000009 WindowServer
03:34:01 CACHE_HIT 0.000013 WindowServer
03:34:01 CACHE_HIT 0.000010 WindowServer
03:34:01 CACHE_HIT 0.000024 WindowServer
03:34:01 CACHE_HIT 0.000010 WindowServer
03:34:01 CACHE_HIT 0.000007 WindowServer
03:34:01 CACHE_HIT 0.000007 WindowServer
03:34:01 CACHE_HIT 0.000007 WindowServer
03:34:01 CACHE_HIT 0.000007 WindowServer
03:34:01 WrData 0.000256 W update
03:34:01 sync 0.258678 W update
03:34:01 WrMeta[async] 0.001092 W update
03:34:01 WrMeta[async] 0.001136 W update
03:34:01 WrMeta[async] 0.001335 W update
03:34:01 WrMeta[async] 0.001508 W update
03:34:01 WrMeta[async] 0.001664 W update
03:34:01 WrMeta[async] 0.001882 W update

I noted though, some things were different (dunno if it's due to the change of name of w):

03:31:31.226 WrMeta[async] D=0x00002778 B=0x1000 /dev/disk0s9 0.001193 W update
03:31:31.226 WrMeta[async] D=0x0000b468 B=0x2000 /dev/disk0s9 0.001366 W update
03:31:31.227 WrMeta[async] D=0x00011878 B=0x2000 /dev/disk0s9 0.001491 W update
03:31:31.227 WrMeta[async] D=0x00016658 B=0x2000 /dev/disk0s9 0.001603 W update
03:31:31.227 WrMeta[async] D=0x0001a0c8 B=0x2000 /dev/disk0s9 0.001754 W update


notice how it says /dev/disk0s9. I believe this means it's doing somethnig to my startup volume.



Hmm... I'm not sure what to do right now. Seems like things are relatively unchanged. Thanks a lot for your help thuh Freak, I really appreciate it. do you think it might be a hardware problem? Cause because this was a fresh install, other people should have it too. I wouldn't assume my comp would be any different. Please reply, thanks again.

[thuh Freak]

i did a little more investigating, and i feel a bit sheepish now. according to the manpage for 'fs_usage', the 'W' we saw wasn't referring to a program, but to the time period on the other side (left side). 'W' means the time includes "wait time". so, you can move the /usr/bin/w.OUTOFTHEWAY back to /usr/bin/w, if you haven't already.

i looked up the 'update' program, and hark it is indeed a very heavy hitting file system call. in modern operating systems, storage media are typically orders of magnitude slower than RAM and the cpu. so, to avoid the horrendous delays when dealing with hard drives and disks, the OS takes a chunk of RAM and pretends like it is the hard drive, and periodically pushes that RAM back to the real physical media. ejecting or unmounting a disk will force this push (and since real data is often in RAM and may not yet be on disk, you'll often get warnings about properly ejecting a disk). well, enough of my rambling.

the 'update' call synchronizes the RAM fake drive and the real physical drive. If you suspect, or know, that some hard data is in RAM and not on disk, you can call 'update'. when it returns, your disks will be updated. it seems that periodically, the OS is calling update, to sync up the disks. two consecutive calls to update, in theory, won't actually touch the hd (the second time) since it should be all sync'd up. but, another program is still playing with the hard drive so its out of date again.

cupsd is the "Common Unix Printing System Daemon". a daemon, in computer science, is a program that runs in the background with no input or ouput. often, and in the case of cups, a server (server, here referring to a piece of software, not hardware) will be a daemon and will accept incoming connections from clients, then "fork" another program to handle all further communication with itself and the client. i don't know too much about the specifics of cups (as to why it would be in use so much, unless you have a document printing, or another computer attempting to access a printer through your computer). but, i know that you can turn it off. i'm more of a command line guy, and from what i remember its not easy or obvious to do what i'm about to describe via the GUI. also, if your printer(s) use cups, you won't want to do the following. iirc, you'll need cups if you are sharing your printer with more than one computer, accessing a printer through another cups, or if your printer directly needs cups, or, of course, if you remember a deliberate requirement somewhere in the specs from your printer for cups. worst case scenario, we notice that you're printer does require cups, and we turn it back on.

in terminal, try the following:
sed 's,\(CUPS=-\).*\(-\),\1NO\2,' /etc/hostconfig > /tmp/newconfig
sudo mv /tmp/newconfig /etc/hostconfig
(c+p the above, to make sure you get the quoting, spacing and greater than symbol correct)

if you had opened /etc/hostconfig before hand, you'd notice that there was a line with "CUPS=-YES-" (or possibly "CUPS=-AUTOMATIC-"), and afterward it will be "CUPS=-NO-". you'll have to restart for it to take effect. after coming up from a restart, i recommend you try out your printer and make sure it works. if it doesn't work, we can revert back with the following (in terminal again).
sed 's,\(CUPS=-\).*\(-\),\1YES\2,' /etc/hostconfig > /tmp/newconfig
sudo mv /tmp/newconfig /etc/hostconfig

assuming that your printer works with cups off, you can try idling again, and see if the hdd spins down. hopefully we're a little closer.

[alcimedes]

you didn't tell it to index your drive, did you?

[stevegong]

you didn't tell it to index your drive, did you?

no, never.

[stevegong]

well thuh Freak, I did exactly as you said but:

12:02:07 WrData 0.000294 W update
12:02:07 sync 0.832036 W update
12:02:07 WrMeta[async] 0.001229 W update
12:02:07 WrMeta[async] 0.001170 W update
12:02:07 WrMeta[async] 0.001275 W update
12:02:07 WrMeta[async] 0.001479 W update
12:02:07 WrMeta[async] 0.001688 W update
12:02:07 WrMeta[async] 0.001890 W update
12:02:07 WrMeta[async] 0.002072 W update
12:02:07 WrMeta[async] 0.002268 W update


:mad: :mad:

I don't know, seems like these hard drive access are much more infrequent. I had to wait almost 5 minutes to get this again in fs_usage. the only problem is that this still won't sustain a hard disk spin down. I woke up this morning still to a hot hard disk :mad:

Do you think we should assume it's a hardware problem at this point? What hardware problem could it possibly be? the power manager?

Thanks so much for your help so far.

so it seems like the w update happens even after the modifications.


just fyi, I don't use a printer with my comp, so it's all good.

any more ideas? thanks.

[thuh Freak]

i suspect that the 10 or so consecutive listings of update within the same second, are from one run of the program. i'd be more interested in knowing if anything is happening between update calls. i'm retissent to call it a hardware problem, b/c whenever i tend to think something is, it turns out i missed something obvious in software.

[Dave]

Maybe the sync is somehow failing without an error message. Then the OS notices that what's on the drive isn't the same as what's in RAM, and tries to sync it again. I have no idea how this could happen though.

Oh, have you tried resetting the Power Manager? I forget how to do it, but somebody here is bound to know.

[stevegong]

Maybe the sync is somehow failing without an error message. Then the OS notices that what's on the drive isn't the same as what's in RAM, and tries to sync it again. I have no idea how this could happen though.

Oh, have you tried resetting the Power Manager? I forget how to do it, but somebody here is bound to know.


You know, that kinda makes sense. I have reset the ppower manager though and it changes nothing. Is it safe to assume at this point that some hardware is broken?

[Barto]

"WrMeta" is almost certainly adaptive hot file clustering (http://www.kernelthread.com/mac/apme/fragmentation/) moving files to your metadata zone.

When a file is opened on an HFS+ volume, the following conditions are tested:

If the file is less than 20 MB in size

If the file is not already busy

If the file is not read-only

If the file has more than eight extents

If the system has been up for at least three minutes

If all of the above conditions are satisfied, the file is relocated -- it is defragmented on-the-fly.

WindowServer probably has an equally innocuous explanation for writing to your hard drive.

Barto

[Barto]

I ran fs_usage on my iBook with everything unplugged. I'm getting a hojillion of those "CACHE_HIT" things a second, from WindowServer, smbclient, xinetd, smbd, MDNSResponder, stuff from DirectoryServer, lookupd, you name it.

It would be nice to understand everything, sometimes you have to just "let it go" though. ;)

[thuh Freak]

I ran fs_usage on my iBook with everything unplugged. I'm getting a hojillion of those "CACHE_HIT" things a second, from WindowServer, smbclient, xinetd, smbd, MDNSResponder, stuff from DirectoryServer, lookupd, you name it.

It would be nice to understand everything, sometimes you have to just "let it go" though. ;)
cache hits, if computer architecute taught me anything (and i did get an a in it), means the hd isn't actually touched. the os keeps an in-RAM cache of popular parts of the hd, and since ram is like a couple of orders of magnitude faster than hds its wicked, crazy, silly, crazy, wicked fast.

[stevegong]

THanks barto for the info, but I'm not sure what to make of it all. What does it mean?